Coronary heart disease (CHD) is the leading cause of death and disability in our society. Most CHD deaths occur in subjects over 70 years of age. Significant independent CHD risk factors are age, gender, elevated low density lipoprotein (LDL) cholesterol (C), decreased high density lipoprotein (HDL) C, hypertension, smoking, diabetes, elevated lipoprotein (a) or Lp(a) (LDL C> 50% reduction), and elevated C-reactive protein. In this response to RFA HL-03-001 (ancillary pharmacogenetic studies), we propose to study 2804 male and 3000 female participants in the Prospective Study of Pravastatin in the Elderly at Risk (PROSPER), who were selected for age 70-82 years, having vascular disease coronary, cerebral or peripheral) or increased CHD risk due to smoking, hypertension or diabetes and total cholesterol levels between 4.0 and 9.0 mml/L or 151 and 340 mg/dl. In this randomized controlled trial pravastatin decreased LDL C 34% and triglyceride 12% and raised HDL C 5%. C-reactive protein and Lp(a) values have already been measured. Fatal and nonfatal myocardial infarction (MI) were decreased by 19%, and fatal MI 24%, but increased risk of new cancer were noted in the pravastatin group over 3.2 years as compared to the placebo group (all p<0.01) (Lancet 360: 1623-30, 2002). Benefit was greatest in subjects with low HDL C (<1.1 lmml/L or 43 mg/dl). No benefit of pravastatin versus placebo on cognitive function or stroke was noted. We and others have shown that statins increase large alpha 1 migrating apolipoprotein A-I containing HDL, decrease plasma lathosterol, a marker of cholesterol synthesis, and increase plasma betasitosterol, a marker of cholesterol absorption as well as decrease cholesterol ester transfer protein (CETP) mass. We propose to measure HDL subspecies, CETP mass, lathosterol, and beta-sitosterol in the 292 subjects who developed CHD while on pravastatin and in a control group (n=292) who did not develop CHD on pravastatin. We propose to isolate DNA in all subjects, carry out sequencing for single nucleotide polymorphism detection in 5 male and 5 female hyper-responders and the same number of hypo-responders (LDL C <10% reduction) and then genotyping at all SNPs on the two 292 patients groups, and the informative SNP detection on the entire 5804 cohort at the following gene loci: ATP binding cassette transporters G5 and G8 (ABCG5, ABCG8), CETP; HMG CoA reductase, apolipoprotein E, lipoprotein and hepatic lipase, microsomal transfer protein, C-reactive protein, connexin, plasminogen activator type I inhibitor and stromelysin I. These genes have been selected because of our own preliminary studies, and their known key role in cholesterol absorption and lipoprotein metabolism or CHD. We hypothesize that response to pravastatin in terms of lowering of LDL C, triglycerides and C-reactive protein, and HDL C raising will be related to specific genotypes and haplotypes. We also hypothesize that subjects with the greatest LDL C- and C-reactive protein-lowering, the greatest increase in large alpha HDL particles, the greatest reduction in lathosterol and the least increase in beta-sitosterol will have the greatest benefit in CHD risk reduction, and that these changes will be related to specific genotypes and haplotypes of the candidate genes being examined. These results can be used to formulate guidelines for identifying elderly subjects for statin treatment to prevent future CHD.